Getty Images
Clockwise from top left: AB de Villiers and Brendon McCullum suffer from opening the innings, while Shane Watson and Tillakaratne Dilshan prosper © Getty Images

It is the general belief that facing fast bowlers with the new ball makes opening more difficult than batting down the order. However, we tend to attach less importance of the value of making first use of a placid track and other perks of an opener’s job. Arunabha Sengupta analyses those batsmen who have opened as well as batted lower in Test cricket and concludes that there is no evidence to say that opening is a more challenging task.

Excerpt from television commentary during a Test match: Two former cricketers — both famed openers in their days—presiding over the microphone. A table on the screen displays the career records of some top scorers of all time.

Commentator 1: “We see some great names there. But I ought to say that when you look at the numbers of batsmen, you can add at least 5 to the averages of the men who played as openers. Right, C2?”

Commentator 2: “Absolutely C1. Opening is the most difficult job.”

And so it goes. Opening is established as the most difficult job in batsmanship. Mostly by mutual consensus.

There is the new ball after all. With all its shine. Fresh bowlers, a lot of them fast. Negotiating the shiny new red cherry up front is a supreme task.

From where did we get the specific estimate of five runs that are to be added to an opener’s average?

Well, er, because it is a nice round number… human beings are born with five fingers, hence there is a fascination for it.

The answer is that ‘five runs’ was derived on the fly, out of the top of the head of an ex-opening batsman who, through a lifetime, has convinced himself that an opener’s job is more difficult than that of a middle-order batsman.

No one expects it to be a precise estimate. What the cricketers mean is that some additional points need to be awarded to an opener’s performance, in order to conduct a fair comparison with the other batsmen.

And it is believed as well, very widely. Most fans feel that an opener should always be given special points because of the difficulties associated with his job.

Has there been any validation of this claim? Is the job of the opener really more difficult than that of, say, a number three or number five batsman?

No, we all just know it is more difficult. Can anyone watch cricket and not realise that facing fresh, fit, fire-breathing fast bowlers with the new ball is way, way more difficult than coming in later and taking on tired bowlers when the ball is older?

This is an illusion known as the Von Restorff effect. The fast bowlers bowling with a new ball is an impactful image, it sticks out like a sore thumb. What gets unnoticed are the occasions when the pitch is a placid shirtfront and the openers get set to bat for hours, perhaps a day or more. On these occasions it is often the middle order men who cannot capitalise, having to wait in the pavilion as runs are piled on by the men who went in first. But that does not create as sensational an image as a Curtly Ambrose running in. Hence, the perks of an opener’s job are generally not perceived as important and a balancing factor.

To me it seems that the number of times an opener is dismissed early by the moving ball hurled down by fresh fast bowlers is more than compensated by the number of occasions they find themselves given the first use of a flat wicket with nothing in it for the bowlers, sometimes even against a toothless attack.

It can be argued that the openers perhaps have to face as much reverse swing with the old ball or as much of the poison tipped offerings of the champion spinners as the middle-order batsman, but it does seem a bit contrived.

How do we scientifically test if openers indeed have a more difficult job or not?

The best way to do this is to take a cohort of batsmen who have played a significant number of innings both as opening batsmen and in other positions in the batting order. And then we can find out whether or not there is a very high proportion of the numbers being boosted by coming down the order as opposed to opening the innings.

For this exercise, we have taken all the batsmen from the history of Test cricket who have played at least 30 innings as an opening batsman as well as at least 30 innings down the order.

In a curious numerical quirk, there have been exactly 30 batsmen in the history of Test cricket who have played at least 30 innings as opener as well as 30 innings lower down the order.

The records of these 30 men are given in the table, categorised according to opener and non-opener.

As we can see of the 30 batsmen, as many as 20 perform better as openers, while 10 suffer a setback.

Comparative records of individual batsmen as opener and non-opener

Player As opener Down the order Diff as opener Effect of Opening
Inns Runs Ave Inns Runs Ave
David Boon (Australia) 63 2614 45.06 127 4808 42.92 2.14 +ve
Mark Butcher (England) 49 1418 29.54 82 2870 37.76 -8.22 -ve
Colin Cowdrey (England) 38 1527 42.41 150 6097 44.5 -2.09 -ve
AB de Villiers (South Africa) 35 1265 36.14 124 6031 55.84 -19.7 -ve
Tillakaratne Dilshan (Sri Lanka) 53 2170 42.54 92 3322 40.02 2.52 +ve
John Edrich (England) 82 3430 44.54 45 1708 41.65 2.89 +ve
Farokh Engineer (India) 48 1577 32.85 39 1034 28.72 4.13 +ve
Grant Flower (Zimbabwe) 84 2373 29.29 39 1084 30.11 -0.82 -ve
Herschelle Gibbs (South Africa) 116 5242 47.22 38 925 25.69 21.53 +ve
Graham Gooch (England) 184 7811 43.88 31 1089 35.12 8.76 +ve
Hanif Mohammad (Pakistan) 65 2638 41.87 32 1277 49.11 -7.24 -ve
Wavell Hinds (West Indies) 46 1482 32.21 34 1126 34.12 -1.91 -ve
Sanath Jayasuriya (Sri Lanka) 152 5932 41.48 36 1041 33.58 7.90 +ve
Simon Katich (Australia) 61 2928 50.48 38 1260 36.00 14.48 +ve
Justin Langer (Australia) 115 5112 48.22 67 2584 40.37 7.85 +ve
Brendon McCullum (New Zealand) 30 969 34.60 121 4337 37.71 -3.11 -ve
Niel McKenzie (South Africa) 35 1279 39.96 59 1974 35.89 4.07 +ve
Majid Khan (Pakistan) 49 1985 42.23 57 1946 36.03 6.20 +ve
Vinoo Mankad (India) 40 1548 40.73 32 561 19.34 21.39 +ve
Bruce Mitchell (South Africa) 48 2390 56.90 32 1081 37.27 19.63 +ve
Ian Redpath (Australia) 59 2492 44.50 61 2245 42.35 2.15 +ve
Wilfred Rhodes (England) 43 1469 36.72 55 856 23.13 13.59 +ve
Bob Simpson (Australia) 70 3664 55.51 41 1205 31.71 23.80 +ve
Alec Stewart (England) 77 3348 44.64 158 5115 36.79 7.85 +ve
Bert Sutcliffe (New Zealand) 41 1763 45.20 35 964 33.24 11.96 +ve
Herbie Taylor (South Africa) 34 1228 36.11 42 1708 44.94 -8.83 -ve
Victor Trumper (Australia) 52 1650 33.00 37 1513 48.80 -15.80 -ve
Michael Vaughan (England) 72 3093 45.48 75 2626 37.51 7.97 +ve
Shane Watson (Australia) 52 2049 40.98 45 1359 30.88 10.1 +ve
Keppler Wessels (Australia/South Africa) 37 1383 39.51 34 1405 42.57 -3.06 -ve

We find the results failing to show us any trend. People like Simon Katich, Herschelle Gibbs, Vinoo Mankad and Bobby Simpson have much better records as openers. At the same time, AB de Villiers and Victor Trumper have much better records down the order.

Some of the statistics are eye-opening. For example, Hanif Mohammad and Trumper, generally considered two of the greatest openers the world has seen, perform significantly better down the order. At the same time, Majid Khan and Neil McKenzie, men who have batted more often down the order, fare better as openers.

Of those faring better as openers, we may encounter slight objections in the study of Graham Gooch, who played only 14% of his innings down the order. But the same objection can be voiced for Brendon McCullum, who has better numbers down the order but played 80% of his innings there. But the rest of the ratios are reasonable enough.

For those who are wondering about the absence of names such as Ravi Shastri and Manoj Prabhakar, they do not satisfy the 30 innings mark.

Shastri played 26 innings as an opener and scored at 44.04, as opposed to 33.28 while coming down the order.

Prabhakar played 30 innings as an opener averaging 35.48, but only 28 as a non-opener in which he scored at 29.18.


  • There is no reason to believe that opening the innings makes it more difficult for a batsman to score runs. As mentioned there are enough balancing factors, which are perhaps not as impactful to the perception as new balls and fierce fast bowlers.
  • A majority of the men in the list boost their records while batting at the top… So, the perception that opening is a more difficult job than coming down in the middle order may actually be rather misleading

There is absolutely no need to add five — or any other number of runs — to the average of an opener. Yes, even if suggested by the mutual agreement of two opening legends.


(Arunabha Sengupta is a cricket historian and Chief Cricket Writer at CricketCountry. He writes about the history and the romance of the game, punctuated often by opinions about modern day cricket, while his post-graduate degree in statistics peeps through in occasional analytical pieces. The author of three novels, he can be followed on Twitter at